Enable per-session setting for CPUs required per task for the jobs in the session.
Enable per-session setting for CPUs required per task for the jobs in the session. Unlike Spark's system-wide setting, this will only pretend to require those many CPUs for scheduler though backend will not mark those many CPUs as busy. The net effect is that lesser number of tasks will be scheduled for the jobs on session and remaining executor cores will still be available for other tasks.
Limit the maximum number of rows in a column batch (applied before ColumnBatchSize property).
Special property to trigger authentication by SnappyConf.
Constant names suggested per naming convention http://docs.scala-lang.org/style/naming-conventions.html
we decided to use upper case with underscore word separator.